Telephone interview was adopted to collect all data required. To minimize sampling bias, telephone numbers were first drawn randomly from the residential telephone directories as "seed numbers", from which another set of numbers was generated using the "plus/minus one/two" method so as to capture those unlisted ones. Duplicated numbers were then filtered out and the remaining numbers were mixed in random order to compile the final telephone sample.
The target population of this survey was mobile phone users in Hong Kong aged 18 or above who spoke Cantonese. When more than one qualified subject was available in a household, the one who would have his/her birthday next was selected. A total of 1,535 respondents were successfully interviewed between 8-20 May 2001. The overall response rate was 66.4%.